7412
J. Am. Chem. Soc. 2000, 122, 7412-7413
Catalytic Enantioselective Cyanosilylation of Ketones
Scheme 1. Synthesis of 1-L
Yoshitaka Hamashima, Motomu Kanai, and Masakatsu Shibasaki* Graduate School of Pharmaceutical Sciences The UniVersity of Tokyo, Hongo, Bunkyo-ku Tokyo, 113-0033, Japan ReceiVed May 13, 2000 The catalytic asymmetric addition of cyanide to carbonyl compounds1 such as aldehydes,2 imines3 and ketoimines4 is currently intensively studied. However, no practical asymmetric cyanosilylation of ketones has been reported so far. For example,5 the best result using chemical catalyst is 72% ee in the case of aryl methyl ketones. However, this catalyst could not be applied to ethyl ketones ( ∼30% ee) and aliphatic ketones.6 In view of the importance of the cyanohydrins as precursors of chiral quaternary R-hydroxy carbonyl derivatives, development of an efficient catalytic asymmetric cyanosilylation of ketones with broad generality is long awaited. Herein, we describe the first entry in this category that we believe is useful for synthesizing a variety of quaternary cyanohydrins, catalyzed by a novel titanium catalyst 1.
Table 1. Effect of Metals
entry
metal
temp/°C
time/h
yield/%
ee/%
R/S
1 2 3 4 5
Et2AlCl Yb(OiPr)3 Zr(OtBu)4 Ti(OiPr)4 Ti(OiPr)4
rt rt rt rt -20
48 2 36 48 36
0 90 52 78 44
18 14 35 73
S R R R
Table 2. Effect of Solvents
During the course of our studies to develop a new asymmetric catalyst from the concept of bifunctional catalysis,7 we have found that the Lewis acid (Al)-Lewis base (phosphine oxide) catalyst 3 can promote the cyanosilylation of acetophenone, however, with low enantiomeric excess (20%).8 To improve the enantioselectivity, we planned to introduce a catechol moiety at the C3 hydroxyl group on the basis of the following consideration. The coordination of the ether oxygen at C3 should make it possible to form a complex such as 1. As a result, the phenyl group of the catechol should be fixed at the position shielding the R-side (anti (1) (a) Gregory, R. J. H. Chem. ReV. 1999, 99, 3649-3682. (b) Effenberger, F. Angew. Chem., Int. Ed. Engl. 1994, 33, 1555-1564. (c) North, M. Synlett 1993, 807-820. (2) For example: (a) Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 2641-2642. (b) Belokon´, Y. N.; CavedaCepas, S.; Green, B.; Ikonnikov, N. S.; Khrustalev, V. N.; Larichev, V. S.; Moscalenko, M. A.; North, M.; Orizu, C.; Tararov, V. I.; Tasinazzo, M.; Timofeeva, G. I.; Yashkina, L. V. J. Am. Chem. Soc. 1999, 121, 3968-3973. (c) Hwang, C.-D.; Hwang, D.-R.; Uang, B.-J. J. Org. Chem. 1998, 63, 67626763. (3) For example: (a) Takamura, M.; Hamashima, Y.; Usuda, H.; Kanai, M.; Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39, 1650-1652. (b) Sigman, M. S.; Vachal, P.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2000, 39, 12791281. (c) Porter, J. R.; Wirschun, W. G.; Kuntz, K. W.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2000, 122, 2657-2658. (d) Ishitani, H.; Komiyama, S.; Hasegawa, Y.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122, 762-766. (e) Corey, E. J.; Grogan, M. J. Org. Lett. 1999, 1, 157-160. (4) (a) Vachal, R.; Jacobsen, E. N. Org. Lett. 2000, 2, 867-870. (b) Byrne, J. J.; Chavarot, M.; Chavant, P.-Y.; Valle´e, Y. Tetrahedron Lett. 2000, 41, 873-876. (5) Belokon´, Y. N.; Green, B.; Ikonnikov, N. S.; North, M.; Tararov, V. I. Tetrahedron Lett. 1999, 40, 8147-8150. (6) Enzymatic reactions have been reported to give cyanohydrins from aliphatic ketones with high enantioselectivity. However, synthesis of cyanohydrins from aromatic ketones and ethyl ketones is not efficient using enzymes. For example: Kiljunen, E.; Kanerva, L. T. Tetrahedron: Asymmetry 1997, 8, 1551-1557. (7) (a) Shibasaki, M. Enantiomer 2000, 4, 513-527. (b) Shibasaki, M. CHEMTRACTS: Org. Chem. 1999, 979-998. (c) Shibasaki, M.; Sasai, H.; Arai, T. Angew. Chem., Int. Ed. Engl. 1997, 36, 1236-1256. (8) Kanai, M.; Hamashima, Y.; Shibasaki, M. Tetrahedron Lett. 2000, 41, 2405-2409.
entry
solvent
conc/M
temp/°C
time/h
yield/%
ee/%
1 2 3 4
CH2Cl2 toluene THF THF
0.65 0.65 0.65 3
-20 -20 -20 -30
36 36 36 36
44 40 58 85
73 70 83 92
to the phosphine oxide, concave side) of the catalyst, thus defining the position of the coordinating ketone at the β-side, syn to the Lewis basic phosphine oxide. Therefore, we designed the new catalyst 1. Ligand 1-L was readily synthesized in multigram scale from the known alcohol 49 as shown in Scheme 1.10 First, we screened different metals combined with ligand 1-L for the catalysis of the addition of TMSCN to acetophenone 9a (Table 1). Although the Yb catalyst showed a remarkable reactivity (Table 1, entry 2), it was found that the Ti catalyst gave the best enantiomeric excess (Table 1, entry 4). Furthermore, when the reaction was conducted at -20 °C, the ee was increased up to 73% (Table 1, entry 5). Next, we examined the effect of solvent (Table 2). Interestingly, both the reaction rate and enantioselectivity increased in a coordinating solvent such as THF compared to less polar solvents such as CH2Cl2 or toluene. Gratifyingly, employing more concentrated conditions (3 M in terms of 9a), the reaction proceeded more efficiently at -30 °C for 36 h to give the product in 85% yield and with 92% ee (Table 2, entry 4). Consequently, the best reaction conditions were determined to involve 10 mol % of Ti(Oi Pr)4 and ligand 1-L in THF solvent. (9) Nakamura, H.; Tejima, S.; Akagi, M. Chem. Pharm. Bull. 1966, 14, 648-657. (10) See Supporting Information.
10.1021/ja001643h CCC: $19.00 © 2000 American Chemical Society Published on Web 07/13/2000
Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 30, 2000 7413
Table 3. Catalytic Asymmetric Cyanosilylation of Ketonesa
Figure 1. Working model
a The method for preparation of the catalyst and the general procedure of the reaction, see ref 12. b Isolated yield. c Determined by chiral HPLC or GC analysis. See Supporting Information. d Ee was determined after conversion to the corresponding benzyl carboxylate. e The absolute configurations were determined by the comparison with the reported values of optical rotation.
To get a preliminary insight into the catalyst structure, we performed NMR studies. When a mixture of Ti(OiPr)4 and 1-L was heated at 75 °C for 1 h in toluene, generation of 2 equiv of iPrOH was observed in 1H NMR. Therefore, at this stage, the pre-catalyst contains titanium diisopropoxide (1: Mtl ) Ti(OiPr)2). After evaporation of toluene, THF and TMSCN (2 equiv to Ti) were added. Then, peaks corresponding to TMSOiPr (0.19, 1.21, and 4.1 ppm) emerged, indicating the generation of titanium cyanide species. After 1 h at ambient temperature, about 70% of titanium seemed to contain monocyanide, deduced from the integration ratio of the remained TMSCN (0.44 ppm) and generated TMSOiPr. Starting the reaction by further adding acetophenone 9a (10 equiv to Ti) and TMSCN (15 equiv to Ti), almost complete mono-ligand exchange from isopropoxide to cyanide seemed to take place.11 Therefore, the actual catalyst should be a complex composed of titanium monocyano monoisopropoxide (1: Mtl ) Ti(CN)(OiPr)). As will be described later, complete formation of titanium monocyanide was realized by longer reaction time (10 h) using 1 equiv of TMSCN. This newly developed catalyst 1 shows a broad applicability for reactions of various ketones with high enantioselectivity, including both aromatic and aliphatic ketones (Table 3).12 (11) For a ligand exchange on titanium from isopropoxide to cyanide by TMSCN, see: Mori, M.; Imma, H.; Nakai, T. Tetrahedron Lett. 1997, 38, 6229-6232. (12) A representative procedure: To a suspension of 1-L (24 mg, 0.0565 mmol) in toluene (1 mL) was added Ti(OiPr)4 (16 µL, 0.054 mmol) at ambient temperature, and the whole was stirred at 75 °C for 1 h. After the yellow solution was cooled to room temperature, toluene was evaporated under reduced pressure. The resulting pale yellow residue was further dried in vacuo for 1 h. The residue was dissolved in THF (0.18 mL, 0.27 mL for aliphatic ketones), TMSCN (14 µL, 0.108 mmol) was added under ice bath, and the whole was stirred at room temperature for 30 min. To this catalyst solution, the starting ketone (0.54 mmol) was added, followed by the addition of TMSCN (144 µL, 1.08 mmol) at the temperature shown in Table 3. The reaction was monitored by TLC, and after the reaction period described in Table 3, pyridine (0.1 mL) and H2O (1 mL) were added. Usual workup and purification by silica gel column chromatography gave the product.
Specifically, less reactive ketones such as propiophenone 9f and indanone 9e gave the product in 89% and 72% yield with 91% and 69% ee’s. The enone 9g gave the 1,2-adduct with complete regioselectivity. Even the simple n-alkanone 9j gave the product in 76% ee.13 Thus, this is the first example of a general cyanosilylation reported to date. The product 10d was successfully converted to the quaternary hydroxy ester (HCl-EtOH, 90 °C for 3 h) or aldehyde (DIBAL-H) in a single step without any loss of enantiomeric excess.10 To get a further insight into the nature of this reaction, kinetic studies were carried out and the reaction rate was found to show a first-order dependency on the catalyst.10 Furthermore, from the labeling experiment using TMS13CN,14 the cyanide appeared to be transferred from TMSCN but not from titanium cyanide.10 Thus, we prepared the active titanium catalyst containing 12CN from Ti(OiPr)4, 1-L (1 equiv) and TMS12CN (1 equiv) (rt for 10 h). After complete consumption of TMS12CN was confirmed by 1H NMR, and then 9a (1 equiv) and TMS13CN (1 equiv) were added. Incorporation (77%) of 13CN into the product 10a was confirmed by 13C NMR.15 These results suggest that the titanium cyanide would act only as a Lewis acid but not as a CN source. Meanwhile, preliminary studies to elucidate the role of the phosphine oxide revealed the importance of this moiety on the enantioselectivity as well as the catalytic activity. Thus, using the control ligand 2 containing a diphenylmethyl group, instead of the phosphine oxide, neither reaction of 9a nor more reactive 9i proceeded at low temperature. These reactions proceeded very slowly at ambient temperature to give 10a and 10i in only 31% and 33% yields (80 h), respectively, and both with 2% ee. From these observations, as well as the former results from our laboratories,2a,3a,7 we propose a dual activation mechanism by the catalyst 1, in which the titanium and the oxygen atom of the phosphine oxide activate the ketone and TMSCN as a Lewis acid and a Lewis base, respectively (Figure 1). In conclusion, we have developed the first general and highly enantioselective cyanosilylation of ketones by designing the novel bifunctional catalyst 1 containing titanium and phosphine oxide. Products were efficiently converted to chiral quaternary R-hydroxy carbonyl derivatives. This contribution should provide a new synthetic strategy for the construction of chiral quaternary carbon centers as well as facilitate biological studies using quaternary R-hydroxy carboxylic acids as a tool.16 Studies for clarifying the reaction mechanism and the origin of the enantioselection are currently under investigation. Acknowledgment. Financial support was provided by CREST, The Japan Science and Technology Corporation (JST), and RFTF of Japan Society for the Promotion of Science. Supporting Information Available: Experimental procedures and characterization of the products (PDF). This material is available free of charge via the Internet at http://pubs.acs.org.
JA001643H (13) The ee was improved to 82% when the reaction was conducted at -60 °C (67% yield for 161 h). (14) TMS13CN was prepared from K13CN following the reported procedure: Reetz, M. T.; Chatziiosifidis, I. Synthesis 1982, 330. (15) Incomplete incorporation of 13CN would stem from a partial scrambling between titanium cyanide and TMS13CN. (16) Quaternary amino acids are often a component of enzyme inhibitors. See refs 4a.
